國立虎尾科技大學 |

Harnessing Deep Learning to Address Climate Change Impacts and Sustainability Challenges.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Harnessing Deep Learning to Address Climate Change Impacts and Sustainability Challenges./
Author:	Sinha, Saumya.
Description:	1 online resource (110 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 85-03, Section: B.
Contained By:	Dissertations Abstracts International85-03B.
Subject:	Computer science. -
Online resource:	click for full text (PQDT)
ISBN:	9798380166485

Harnessing Deep Learning to Address Climate Change Impacts and Sustainability Challenges.
Sinha, Saumya.

Harnessing Deep Learning to Address Climate Change Impacts and Sustainability Challenges. - 1 online resource (110 pages)

Source: Dissertations Abstracts International, Volume: 85-03, Section: B.

Thesis (Ph.D.)--University of Colorado at Boulder, 2023.

Includes bibliographical references

Climate change has already had several visible effects around the world and is increasing rapidly over time. The changing climate poses numerous threats to our environment. In these challenging times, machine learning scientists have stepped forward in developing mitigation solutions to climate change impacts and devising strategies to adapt to unavoidable changes. UN's top sustainable development goals for the year 2030 are focused on tackling poverty, hunger, climate change, and environmental degradation. Through my research, I aim to investigate urgent problems in the area of climate and sustainability and build data and machine learning (ML) based approaches for addressing these problems. This thesis contains contributions to three such critical problems. In the first chapter of this thesis, I discuss my work on extreme event detection, where I build an unsupervised autoencoder-based solution for detecting extreme events (specifically avalanches in French Alps) using satellite imagery. Developing algorithms for detecting such extreme events can be crucial to mitigating disasters and providing insights into climate change. In the second chapter, I turn to long-term solar energy forecasting and develop deep sequence learning models that also provide uncertainty estimates with the forecast via probabilistic prediction. This is crucial given the increasing need to enable widespread integration of renewable energy sources into our power system. In the third and final chapter, I design a machine learning pipeline to generate sea level rise projections 30 years into the future. To understand the extent of causal contributions from the climate-change signals such as greenhouse gases and aerosols in the sea level change, physics-driven global climate models are included in this study along with altimeter observations. Following a brief survey of machine learning techniques addressing the above broad-level problems, I will identify some of the challenges associated with the current approaches, room for improvements, and describe my research that leverages the recent advancements in AI to tackle the existing limitations and contribute solutions and insights to such high-impact real-world problems.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798380166485Subjects--Topical Terms:

573171
Computer science.
Subjects--Index Terms:

Machine learningIndex Terms--Genre/Form:

554714
Electronic books.

Harnessing Deep Learning to Address Climate Change Impacts and Sustainability Challenges.
LDR:03604ntm a22003857 4500 001 1148470
005 20240924101521.5
006 m o d
007 cr bn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798380166485
035 $a (MiAaPQ)AAI30631326
035 $a AAI30631326
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Sinha, Saumya. $3 1474436
245 1 0 $a Harnessing Deep Learning to Address Climate Change Impacts and Sustainability Challenges.
264 0 $c 2023
300 $a 1 online resource (110 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-03, Section: B.
500 $a Advisor: Monteleoni, Claire.
502 $a Thesis (Ph.D.)--University of Colorado at Boulder, 2023.
504 $a Includes bibliographical references
520 $a Climate change has already had several visible effects around the world and is increasing rapidly over time. The changing climate poses numerous threats to our environment. In these challenging times, machine learning scientists have stepped forward in developing mitigation solutions to climate change impacts and devising strategies to adapt to unavoidable changes. UN's top sustainable development goals for the year 2030 are focused on tackling poverty, hunger, climate change, and environmental degradation. Through my research, I aim to investigate urgent problems in the area of climate and sustainability and build data and machine learning (ML) based approaches for addressing these problems. This thesis contains contributions to three such critical problems. In the first chapter of this thesis, I discuss my work on extreme event detection, where I build an unsupervised autoencoder-based solution for detecting extreme events (specifically avalanches in French Alps) using satellite imagery. Developing algorithms for detecting such extreme events can be crucial to mitigating disasters and providing insights into climate change. In the second chapter, I turn to long-term solar energy forecasting and develop deep sequence learning models that also provide uncertainty estimates with the forecast via probabilistic prediction. This is crucial given the increasing need to enable widespread integration of renewable energy sources into our power system. In the third and final chapter, I design a machine learning pipeline to generate sea level rise projections 30 years into the future. To understand the extent of causal contributions from the climate-change signals such as greenhouse gases and aerosols in the sea level change, physics-driven global climate models are included in this study along with altimeter observations. Following a brief survey of machine learning techniques addressing the above broad-level problems, I will identify some of the challenges associated with the current approaches, room for improvements, and describe my research that leverages the recent advancements in AI to tackle the existing limitations and contribute solutions and insights to such high-impact real-world problems.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Climate change. $2 bicssc $3 1009004
653 $a Machine learning
653 $a Deep learning
653 $a Sustainability challenges
653 $a Numerous threats
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
690 $a 0404
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a University of Colorado at Boulder. $b Computer Science. $3 1180184
773 0 $t Dissertations Abstracts International $g 85-03B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30631326 $z click for full text (PQDT)